Transformer-Based Feature Compensation and Aggregation for DeepFake Detection
Abstract
Deepfake detection has attracted increasing attention in recent years. In this paper, we propose a transformer-based framework with feature compensation and aggregation (Trans-FCA) to extract rich forgery cues for deepfake detection. To compensate local features for transformers, we propose a Locality Compensation Block (LCB) containing a Global-Local Cross-Attention (GLCA) to attentively fuse global transformer features and local convolutional features. To aggregate features of all layers for capturing comprehensive and various fake flaws, we propose Multi-head Clustering Projection (MCP) and Frequency-guided Fusion Module (FFM), where the MCP attentively reduces redundant features into a few concentrated clusters, and the FFM interacts all clustered features under the guidance of frequency cues. In Trans-FCA, besides global cues captured by transformer architecture, local details and rich forgery defects are also captured using the proposed fetaure compensation and aggregation. Extensive experiments show our method outperforms the state-of-the-art methods on both intra-dataset and cross-dataset testings (with AUCs of 99.85% on FaceForensics++ and 78.57% on Celeb-DF), which clearly demonstrates the superiority of our Trans-FCA for deepfake detection.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Cites
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 21,793 · see citing works
Celeb-DF: A Large-Scale Challenging Dataset for DeepFake Forensics
Li, Y., Yang, X., Sun, P., Qi, H., & Lyu, S. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00327 · Cited by 1,846 · cited 1× · see citing works
“Besides, we also employ other three datasets for cross-dataset testing including Celeb-DF (V2) [28], TIMIT [29] (including two settings of low-quality (LQ) and high-quality(HQ)) and UADFV [30] datasets.”
MethodMulti-attentional Deepfake Detection
Zhao, H., Wei, T., Zhou, W., Zhang, W., Chen, D., & Yu, N. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00222 · Cited by 1,019 · see citing works
Spatial-Phase Shallow Learning: Rethinking Face Forgery Detection in Frequency Domain
Liu, H., Li, X., Zhou, W., Chen, Y., He, Y., Xue, H., Zhang, W., & Yu, N. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00083 · Cited by 600 · see citing works
DeepFakes: a New Threat to Face Recognition? Assessment and Detection
Korshunov, P., & Marcel, S. (2018) · arXiv · doi:10.48550/arxiv.1812.08685 · Cited by 497 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
F <sup>2</sup> Trans: High-Frequency Fine-Grained Transformer for Face Forgery Detection
Miao, C., Tan, Z., Chu, Q., Liu, H., Hu, H., & Yu, N. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3233774 · Cited by 172 · see citing works
Self-citationExploring autonomous methods for deepfake detection: A detailed survey on techniques and evaluation
Sunil, R., Mer, P., Diwan, A., Mahadeva, R., & Sharma, A. (2025) · Heliyon · doi:10.1016/j.heliyon.2025.e42273 · Cited by 54 · see citing works
Exploiting Modality-Specific Features for Multi-Modal Manipulation Detection and Grounding
Wang, J., Liu, B., Miao, C., Zhao, Z., Zhuang, W., Chu, Q., & Yu, N. (2024) · ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp48485.2024.10448385 · Cited by 12 · see citing works
“Consequently, this threat has garnered significant attention in the fields of computer vision and natural language processing, leading to the proposal of various methods for detecting fake faces and AI-generated text [3, 4, 5, 6, 7, 8, 9, 10].”
BackgroundSelf-citationBeyond Accuracy: A Deployment-Oriented Benchmark of Deepfake Detection Models
Ain, Q. U., Ning, H., Philipo, A. G., Daneshmand, M., & Ding, J. (2025) · Preprint (TechRxiv) · doi:10.36227/techrxiv.176315939.91577661/v1 · Cited by 2 · see citing works
- Authors
- Tan, Z., Yang, Z., Miao, C., & Guo, G.
- Affiliations
- Baidu (China) (CN)
- Venue
- IEEE Signal Processing Letters
- Published
- 2022
- Volume
- 29
- Pages
- 2183-2187
- ISSN
- 1070-9908 (ISSN-L)
- 1558-2361
- OpenAlex
- W4312732149
- Cited by
- 59
- References
- 5
- Citations by year
- 2026: 10 · 2025: 26 · 2024: 21 · 2023: 2
- FWCI
- 3.87
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Signal Processing Letters (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Keywords
- deepfake detection, transformer models, feature compensation, feature aggregation, cross-dataset generalization
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
